Risk Assessment and Screening for Preterm Birth in Multiple Pregnancy
Amanda Roman, Alexandra Ramirez, Guillermo Gurza and Vincenzo Berghella
Introduction
Preterm birth (PTB) is defined as delivery before 37 completed weeks of gestation, with extreme PTB defined as occurring at less than 28 weeks, very preterm delivery occurring between 28 and 32 weeks and moderate-to-late PTB occurring from 32 to 36 weeks.1 Worldwide, about 15 million babies are born preterm each year, and PTB accounts for 1 million neonatal deaths and an additional 125,000 deaths in children before five years of age, representing the leading cause of both neonatal and childhood mortality.2,3 Prematurity is associated with increased perinatal morbidity and long-term neurodevelop- mental impairment.
Survival and neurological prognoses improve with advancing gestational age. In the United States, the rate of twin births in 2018 was 32.6 per 1,000 births, but they accounted for 20% of all PTB, with 60.3% delivering before 37 weeks, 19.5% delivering before 34 weeks and 10.7% delivering before 32 weeks. Of singleton pregnancies, 8.2% delivered before 37 weeks, 2.1% delivered before 34 weeks and 1.2% delivered before 32 weeks.4 Twins are also at increased risk for low birthweight (LBW) and have a five times higher risk of early neonatal and infant death and complications related to prematurity and LBW.4Pathophysiology
The increased incidence of PTB among twin/multiple pregnancies is probably associated with different pathophysiology: maternal complications associated with physiological changes (cardiovascular and endocrine adaptation to multiple pregnancy), fetal complications unique to multiple pregnancy and directly associated with chorionicity, uterine over-distension, cervical insufficiency, intrauterine infection or inflammation, hormonal disorders, placental insufficiency and uterine hypoxia.
It is therefore important to individualise the risk factors associated with PTB among asymptomatic and symptomatic twin pregnancies.Prevention of Multiple Pregnancy
Avoidance of Multiple Pregnancy
The best practice to prevent PTB is to avoid multiple pregnancy. Diverse societies have recommended a single embryo transfer during in vitro fertilisation (IVF) to avoid multiple pregnancy. However, the twin pregnancy rate remains high among women undergoing artificial reproductive techniques (ART). Approximately 30.4% of ART-conceived infants are twins and 1.1% are triplets and higher-order multiples. The number of fetuses correlates directly with the number of embryos, zygotes or oocytes transferred following reproductive techniques. In a systematic review and meta-analysis, the risk for monozygotic twin (MZT) and monochorionic twin (MCT) pregnancies after IVF is increased with blastocyst transfer (prolonged embryo culture) compared with cleavage-stage embryo transfer (23 studies, OR 2.16, 95% CI, 1.74-2.68, I2 = 78%).5 Conventional IVF compared with intra-cytoplasmic sperm injection (ICSI) and assisted hatching was associated with a statistically significant increased risk of MZT pregnancy (9 studies, OR 1.19, 95% CI, 1.04-1.35, I2 = 0; 16 studies, OR 1.17, 95% CI, 1.09-1.27, I2 = 29%, respectively). Embryo biopsy for pre-implantation genetics, embryo cryopreservation and oocytes donation were not associated with MZT pregnancies after IVF. Prevention of spontaneous conception is not possible, and medical ovarian stimulation and intrauterine sperm transfer still offer a great risk of twins and multiple pregnancy. A single embryo transfer offers the best practice to prevent twins or multiple pregnancy.
Multifetal Reduction and Selective Fetal Termination
Another method to decrease the risk of PTB and neonatal complications in multiple pregnancy is multifetal reduction. A systematic review evaluated embryonic reduction at 8-14 weeks of gestation in both trichorionic-triamniotic (TCTA) and dichorionic- triamniotic (DCTA) pregnancies compared with expectant management (this review included eight studies with a total of 249 DCTA and 1,167 TCTA pregnancies).6 In TCTA pregnancies, there were lower risks (17.3% vs 50.2%) of PTB < 34 weeks (RR = 0.36, 95% CI: 0.28-0.48), whereas the risk of miscarriage < 24 weeks (8.1% vs 7.4%) did not significantly increase (RR = 1.08, 95% CI: 0.58-1.98).
The embryonic reduction to twins in DCTA triplets, either of the fetus with a separate placenta or one of the MC pair, was significantly associated neither with an increased risk of miscarriage (8.5 vs 13.3%, P = 0.628 and RR = 1.22, 95% CI: 0.38-3.95, respectively) nor with a lower risk of PTB (51.9 vs 46.2%, P = 0.778 and RR = 0.5, 95% CI: 0.04-5.7, respectively).Fetal reduction of DCDA twins to a singleton pregnancy before 14 weeks is associated with a lower risk of PTB < 34 weeks (1.6% vs 11.7%) and < 37 weeks (9.5% vs 56.7%). There were also improvement in rates of miscarriage of one twin (0% vs 4.8%) and early pregnancy loss < 24 weeks of gestation, while gestational diabetes (11.1% vs 10%), hypertensive diseases of pregnancy (6.3% vs 15%) and intrauterine growth restriction (0% vs 3.3%) were not significantly different. Early fetal reductions (11-14 weeks) have better outcomes than later fetal reductions (15-23 weeks): preterm delivery < 37 weeks (14% vs 28%, p = 0.004), < 34 weeks (1.8% vs 12%, p = 0.001), < 32 weeks (1.8% vs 8%, p = 0.012) and decreased composite neonatal morbidity (2.9% vs 10.7%, p = 0.025).6
Therefore, fetal reduction should be considered in certain cases of twin and multiple order pregnancy where the risk for adverse outcome seems exceptionally high. The benefit of fetal reduction not only includes lower frequencies of pregnancy loss, PTB and LBW infants, but it also may decrease antenatal complications, caesarean delivery and neonatal deaths.
Chorionicity is an important factor associated with the incidence of PTB and perinatal outcomes in twin pregnancies. Monochorionic-diamniotic (MCDA) twin pregnancies are at higher risk of PTB associated with maternal and fetal complications compared with DCDA twin pregnancies. In a retrospective cohort of 175 (30.2%) MCDA twin pregnancies and 405 (69.8%) DCDA pregnancies, the incidence of PTB < 34 weeks was significantly increased in the MCDA pregnancies: 61 (34.9%) versus 93 (23.0%), respectively with OR 1.80 (1.22-2.65).
the increase was seen both for the medically indicated (fetal reasons) PTB: 22 (12.6%) versus 20 (4.9%) with OR 2.49 (1.34-5.63) and spontaneous PTB (SPTB): 108 (61.7%) versus 195 (48.1%) with OR 1.74 (1.21-2.49), while maternal indication of PTB was not different: 30 (17.1%) versus 65 (16.1%) with OR: 1.05 (0.55-1.70).7Fetal Complications As a Cause of Preterm Birth
Twin-Twin Transfusion Syndrome (TTTS)
A unique complication of MCDA pregnancies is twin-twin transfusion syndrome (TTTS). Twin-twin transfusion syndrome complicates 8-10% of MCDA pregnancies. The prevalence of TTTS is approximately 1-3 per 10,000 births. Twin-twin transfusion syndrome contributes highly to the fetal indications of PTB; a systematic review including 26 studies (2,699 twin pregnancies) showed that the pooled proportions of PTB < 32 weeks of gestation among MCDA pregnancies affected by TTTS classified by Quintero stages were increased as the severity of TTTS progressed. Quintero stage I: two studies (9/34); 27.1% (13.9-42.8) stage II, three studies (20/47); 42.8% (29.4-56.9) stage III, three studies (32/58); 53.3% (36,1-70.2) and stage IV: three studies (11/18) 59.9% (37.9-80). There are also differences in perinatal outcomes according to TTTS treatment modality. Overall PTB occurred in 74.1% (95% confidence interval 36.9-97.8) of pregnancies managed expectantly, 43.5% (26.5-51.3) treated with laser and 65.3% (45.8-80.7) managed with amnioreduction.8 However, data on PTB outcomes were limited by the small sample size.
Other Fetal Complications
Other MCDA-specific complications affecting prematurity rates include selective intrauterine growth restriction (sIUGR) and twin anaemia polycythaemia sequence (TAPS). The gestational age at delivery in 134 MCDA twin pregnancies diagnosed with sIUGR between 18 and 26 weeks was 35.4 (16-38) weeks if umbilical artery (UA) Doppler with positive diastolic flow, and 30.7 (27-40) weeks if persistent absent or reversed end-diastolic flow.
In women affected by TAPS, the rate of preterm premature rupture of the membranes (PPROM) was 44.4% in the treated TAPS group versus 18.2% in the non-treated TAPS group, and the mean gestational age at delivery was 33.4 ± 3 weeks and 30.4 ± 5.7 weeks, respectively.9Single Intrauterine Death in Twin and Preterm Births
Single intrauterine death (sIUD) in twin pregnancy is associated with a significant risk of cotwin demise and PTB, especially in MC twins. The risk of PTB in twin pregnancy after sIUD according to the gestational age at death was evaluated among 3,013 twin gestations (2,469 DC and 544 MC). Median gestational age at birth was lower in the pregnancies complicated by sIUD compared with those that were not (32 weeks: interquartile range (IQR), 29-34.3 weeks vs 36.7 weeks: IQR, 35-37.6; P < 0.001), and this difference persisted when stratifying the data according to chorionicity (P < 0.0001 for both MC and DC pregnancies). The overall risk was higher in pregnancies complicated by sIUD compared with those which did not experience fetal loss. Preterm birth at < 34 weeks was 66.1% (41/62) versus 15.5% (457/2,948), RR 4.27 (3.5-5.2); PTB < 32 weeks 48.4% (30/62) versus 7.9% (233/2,948), RR 6.12 (4.6-8.1); and PTB < 28 weeks 19.4% (12/62) versus 1.6% (46/2948), RR 12.40 (6.922.2). This association was observed both in MC and DC twin gestations. The risk of PTB at < 34 weeks of gestation was higher when the sIUD occurred at a later gestational age (chisquare test for trend, P < 0.001).10 In summary, monochorionicity is a strong risk factor for PTB compared to dichorionicity in twins due to the inherited unique complications such as TTTS, TAPS, single IUFD and selective IUGR.
Maternal Indications for Preterm Birth
History of Previous Preterm Birth
The incidence of history of PTB and subsequent twin pregnancies is about 10% in most studies.7 Several retrospective cohorts have shown an increased risk of PTB in twin pregnancy with history of singleton PTB.
The preterm prediction study in twins showed an increased risk of recurrent PTB < 37 weeks (80% vs 50%; RR 1.6. 95% CI 1.18-2.17, p = 0.03) (n = 147). Three retrospective studies in the United States showed an increased risk of recurrent PTB < 37 weeks in subsequent twin pregnancy (73.9% vs 44.4%; OR: 3.5, 95% CI 1.4-9.37), recurrent spontaneous PTB < 35 weeks (37.7% vs 23.0%; OR 2.0, 95%CI 1.41-2.9, P < 0.001) and recurrent PTB < 32 weeks (26% vs 3.5%; OR 9.7, 95%CI 2.95-32.4, Pgestations.15 Multiple variables were evaluated in 147 women with twin pregnancy at 24 and 28 weeks, including maternal demographics, obstetric history, biochemical markers and transvaginal ultrasound at 24 and 28 weeks. A transvaginal ultrasound cervical length (TVUCL) ≤ 25 mm at 24 weeks was the best predictor of SPTB at < 32, < 35 and < 37 weeks. Since then, multiple prospective studies evaluating TVUCL in twins have been published. A systematic review including 16 studies in asymptomatic women with twin pregnancies (n = 3213) showed that TVUCL ≤ 20 mm at 20-24 weeks was the most accurate tool in predicting PTB < 32 and < 34 weeks of gestation (pooled sensitivities, specificities and positive and negative likelihood ratios of 39% and 29%, 96% and 97%, 10.1 and 9.0, and 0.64 and 0.74, respectively). A TVUCL ≤ 25 mm at 20-24 weeks of gestation had a pooled positive likelihood ratio of 9.6 to predict PTB < 28 weeks of gestation. The predictive accuracy of TVUCL for PTB was low in symptomatic women. Despite large variation in gestational age at TVUCL and different cut-off points for cervical length and definitions of PTB, the summary receiver operating characteristic (ROC) curve indicated a good predictive capacity of short cervical length for PTB. Sensitivity and specificity for PTB < 34 weeks gestation were 78% and 66%, respectively, for 35 mm, 41% and 87% for 30 mm, 36% and 94% for 25 mm, and 30% and 94% for 20 mm.A recent individual patient-level meta-analysis evaluated the effect of gestational age and cervical length measurements in the prediction of SPTB in twin pregnancies.16 A total of 6,188 TVUCL measurements were performed on 4,409 twin pregnancies in 12 studies. Multinomial logistic regression analysis determined probabilities for birth at ≤ 28, 28-32, 32-36 and ≥ 36 weeks as a function of gestational age (GA) at screening and CL measurements. Both GA at screening and TVUCL had a significant and non-linear effect on GA at birth. The best prediction of PTB ≤ 28 weeks was provided by screening at ≤ 18 weeks (P < 0.001) whereas the best prediction of birth between 28 and 36 weeks was provided by screening at ≥ 24 weeks (P < 0.001).
In summary, all these studies established that TVUCL before 24 weeks is the best predictor of PTB in twin pregnancy, and the shorter the TVUCL and the earlier GA at presentation, the higher the risk of PTB at any given gestation. While TVUCL is currently the best tool available to screen for PTB in twin pregnancy, independent of other risk factors, the ACOG and the SMFM still recommend against TVUCL in twins as there are no current proven therapies to offer and TVUCL should be reserved for randomised controlled studies.
Amniotic Fluid Sludge
Amniotic fluid (AF) was described in 2005 as a dense aggregate of particulate matter seen in close proximity to the internal cervical os and having imaging characteristics similar to gallbladder sludge seen by ultrasound. Sludge has been proposed as an additional ultrasound marker for preterm delivery.
In a large cohort including 635 twin pregnancies with TVUCL from 22 0/7 to 25 6/7 weeks, sludge was present in 61 (9.6%) of them. The risk of PTB < 35 weeks in women with sludge was 36.8% versus 16.5% (OR: 2.11; 95% CI: 1.04-4.27). Sludge constitutes the third marker associated with PTB after TVUCL < 25 mm and fetal fibronectin.17 In summary, amniotic fluid sludge is an independent marker of PTB; however, its presence has not been used to select a better candidate in clinical practice.
Cervical Examination
This study evaluated the risk of antepartum cervical examination at every prenatal visit in twin pregnancy from 1988 through 1991. Eighty-nine women received extensive PTB prevention education and routine cervical examination at each clinic visit (7.6 ± 3.2 cervical examinations per patient) versus 288 women who received cervical examination for obstetric indications only. There were no significant differences in PTB < 37 weeks 77.3% versus 66.7%, p = 0.07, medical complications or infectious morbidity. There were significantly less PPROM among the patients attending the Twin Clinic (12.4% vs 23.6%, P = 0.03).18 Larger studies will be needed to assess this finding. In summary, routine manual cervical examinations for prediction or prevention of PTB in twin pregnancies cannot be recommended.
Uterine Activity Evaluation
In 1995, a meta-analysis of six randomised trials involving 260 twin pregnancies compared daily home uterine activity monitoring (HUAM) versus abdominal palpation. This study did not find any difference in the incidence of PTB (RR: 1.01; 95% CI: 0.79-1.30; p = 0.95). However, when women with twin pregnancy and cervical dilation > 2 cm were evaluated, the incidence of preterm labour was significantly decreased by 56%; RR 0.44 (0.25-0.78, p = 0.005).19 In summary, HUAM cannot be recommended for twin pregnancies.
Fetal Fibronectin
A recent systematic review showed that positive fetal fibronectin (fFN) was associated with a significantly increased risk of PTB among women with twin pregnancies who are either asymptomatic or symptomatic for PTB.20 Among symptomatic women, positive fFN was predictive of PTB < 28 weeks (OR 12.06, 95% CI 4.90-29.70, I2 = 0%), PTB recommended
Key Points
• Twin pregnancies have an increased risk of pregnancy loss, PTB and adverse neonatal outcomes.
• The best practice to prevent PTB is to avoid multiple pregnancies. Fetal reduction should be considered in certain cases of twin and higher-order pregnancies where the risk for adverse outcome seems exceptionally high.
• Determining chorionicity during the first trimester is very important in counselling about risk of PTB and perinatal outcomes in twin pregnancies.
• Fetal complications unique to twin pregnancy (TTTS, selective intrauterine sIUGR, TAPS and sIUD) are important indications for PTB.
• Spontaneous labour is the most common cause of preterm delivery. Transvaginal cervical length before 24 weeks is the best tool to predict spontaneous PTB, independent of other risk factors.
• Maternal indications for PTB include pre-eclampsia, gestational diabetes, smoking, age and race.
References
1. Blencowe H, Cousens S, Oestergaard MZ et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet 2012 Jun 9;379 (9832):2162-72. PubMed PMID: 22682464.
2. Nour NM. Premature delivery and the millennium development goal. Rev Obstet and Gynecol 2012;5(2):100-5. PubMed PMID: 22866189. Pubmed Central PMCID: 3410509.
3. Liu L, Oza S, Hogan D et al. Global, regional, and national causes of under-5 mortality in 2000-15: an updated systematic analysis with implications for the Sustainable Development Goals. Lancet 2016 Dec 17;388 (10063):3027-35. PubMed PMID: 27839855. Pubmed Central PMCID: 5161777.
4. Martin JA, Hamilton BE, Osterman MJK. Births in the United States, 2018. NCHS data brief. 2019 Jul(346):1-8. PubMed PMID: 31442195.
5. Busnelli A, Dallagiovanna C, Reschini M, Paffoni A, Fedele L, Somigliana E. Risk factors for monozygotic twinning after in vitro fertilization: a systematic review and meta-analysis. Fertil Steril 2019 Feb;111(2):302-17. PubMed PMID: 30691632.
6. Anthoulakis C, Dagklis T, Mamopoulos A, Athanasiadis A. Risks of miscarriage or preterm delivery in trichorionic and dichorionic triplet pregnancies with embryo reduction versus expectant management: a systematic review and meta-analysis. Hum Repr. 2017 Jun 1;32 (6):1351-9. PubMed PMID: 28444191.
7. Roman A, Saccone G, Dude CM et al. Midtrimester transvaginal ultrasound cervical length screening for spontaneous preterm birth in diamniotic twin pregnancies according to chorionicity. Eur J Obstet Gynecol Reprod Biol 2018 Oct;229:57-63. PubMed PMID: 30107361.
8. D'Antonio F, Benlioglu C, Sileo FG et al. Perinatal outcomes of twin pregnancies affected by early twin-twin transfusion syndrome: A systematic review and metaanalysis. Acta obstetricia et gynecologica Scandinavica 2020 Mar 11. PubMed PMID: 32162305.
9. Gratacos E, Lewi L, Munoz B et al. A classification system for selective intrauterine growth restriction in monochorionic pregnancies according to umbilical artery Doppler flow in the smaller twin. Ultrasound Obstet Gynecol 2007 Jul;30(1):28-34. PubMed PMID: 17542039.
10. D’Antonio F, Thilaganathan B, Dias T, Khalil A, Southwest Thames Obstetric Research Center. Influence of chorionicity and gestational age at single fetal loss on risk of preterm birth in twin pregnancy: analysis of STORK multiple pregnancy cohort. Ultrasound Obstet Gynecol 2017 Dec;50(6):723-7. PubMed PMID: 28150444.
11. Tarter JG, Khoury A, Barton JR, Jacques DL, Sibai BM. Demographic and obstetric factors influencing pregnancy outcome in twin gestations. Am J Obstet Gynecol 2002 May;186(5):910-12. PubMed PMID: 12015510.
12. Wisborg K, Henriksen TB, Secher NJ. Maternal smoking and gestational age in twin pregnancies. Acta obstetricia et gynecologica Scandinavica 2001 Oct;80 (10):926-30. PubMed PMID: 11580737.
13. Henry DE, McElrath TF, Smith NA. Preterm severe preeclampsia in singleton and twin pregnancies. J Perinatol 2013 Feb;33(2):94-7. PubMed PMID: 22678139.
14. Moini A, Shiva M, Arabipoor A, Hosseini R, Chehrazi M, Sadeghi M. Obstetric and neonatal outcomes of twin pregnancies conceived by assisted reproductive technology compared with twin pregnancies conceived spontaneously: a prospective follow-up study. Eur J Obstet Gynecol Reprod Biol 2012 Nov;165 (1):29-32. PubMed PMID: 22884795.
15. Goldenberg RL, Iams JD, Miodovnik M et al. The preterm prediction study: risk factors in twin gestations. National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network.Am J Obstet Gynecol 1996 Oct;175 (4 Pt 1):1047-53. PubMed PMID: 8885774.
16. Kindinger LM, Poon LC, Cacciatore S et al. The effect of gestational age and cervical length measurements in the prediction of spontaneous preterm birth in twin pregnancies: an individual patient level meta-analysis. BJOG 2016 May;123(6):877-84. PubMed PMID: 26333191.
17. Spiegelman J, Booker W, Gupta S et al. The independent association of a short cervix, positive fetal fibronectin, amniotic fluid sludge, and cervical funneling with spontaneous preterm birth in twin pregnancies. Am J Perinatol 2016 Oct;33 (12):1159-64. PubMed PMID: 27434692.
18. Bivins HA Jr, Newman RB, Ellings JM, Hulsey TC, Keenan A. Risks of antepartum cervical examination in multifetal gestations. Am J Obstet Gynecol 1993 Jul;169(1):22-5. PubMed PMID: 8333461.
19. Colton T, Kayne HL, Zhang Y, Heeren T. A metaanalysis of home uterine activity monitoring. Am J Obstet Gynecol 1995 Nov;173(5):1499-1505. PubMed PMID: 7503191.
20. Marleen S, Dias C, MacGregor R et al. Biochemical predictors of preterm birth in twin pregnancies: a systematic review involving 6077 twin pregnancies. Eur
J Obstet Gynecol Reprod Biol 2020 May 5;250:130-42. PubMed PMID: 32446146.